This is the mail archive of the gcc-help@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Bug in while() intrinsic ?


BERTRAND Joel <joel.bertrand@systella.fr> writes:

> Ian Lance Taylor wrote:
>> BERTRAND Joel <joel.bertrand@systella.fr> writes:
>>
>>> 	I'm trying to debug a subroutine in a very long program. You can find
>>> this GPL'd program at http://www.rpl2.net/.
>>
>> It's unlikely that anybody here is going to download and debug your
>> large program.  We do better with standlone test cases.  I can assure
>> you that the while statement is not broken.  I recommend that you step
>> through the code in a debugger and find out what is happening.  Sorry
>> I can't be of more help.
>
> 	If I run it in a debugger (gdb) without any recompilation, it
> _works_ ! I don't look for someone that debugs this program. I only
> give you an example to reproduce this bug. Before my first post, I
> have tried a lot of different configurations (do {} while(), if ()
> break and some others without any differences.). I have tried to debug
> asmb output but if I know sparc and alpha mnemonics, I don't know
> amd64 ones...

A program which fails normally but works when run under the debugger
does not indicate a compiler error.  If there are multiple threads,
the most likely problem is a synchronization error.  Otherwise the
most likely problem is memory corruption, typically due to a stray
pointer.

Ian


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]